Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Graph Kernel Based Social Network Link Prediction Method

A technology of social network and prediction method, which is applied in the field of social network analysis and can solve the problems of insufficient utilization of information on joint network structure

Active Publication Date: 2020-05-05
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Based on this, in order to solve the problem of insufficient utilization of the contact network structure information in the existing work in link prediction, we propose a social network link prediction method based on graph kernel

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Graph Kernel Based Social Network Link Prediction Method
  • Graph Kernel Based Social Network Link Prediction Method
  • Graph Kernel Based Social Network Link Prediction Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 2

[0053] Embodiment 2 of the present invention introduces a method for generating a subnet set for a node, and the specific steps include:

[0054] A. In the positive link network pos_G, with the node v to be processed i , (i in Indicates that in the network pos_G to node v i The network of all nodes whose shortest path is less than t, The edge of is corresponding to the edge that appears in pos_G; where |V| represents the number of all nodes in G, and T represents the determined threshold;

[0055] B. In the negative link network neg_G, with the node v to be processed i , (i in Indicates to node v in network neg_G i The network of all nodes whose shortest path is less than t, The side of is corresponding to the side that appears in neg_G;

[0056] C. In the link network lin_G, with the node v to be processed i , (i in Indicates that in the network lin_G to node v i The network of all nodes whose shortest path is less than t, The side of is corresponding to the ...

Embodiment 3

[0059] Embodiment 3 of the present invention has introduced the method for computing node similarity of graph kernel, and specific steps comprise:

[0060] A. Analyze any pair of nodes v i , v j The similarity between, take the corresponding sub-network under the sub-network set, which generates G 1 ,G 2 The link networks and thresholds of are the same, that is, the links to generate two networks must come from one of the three networks in step A, and the threshold t corresponding to the generated sub-networks must be equal;

[0061] B. Calculate the corresponding power iteration space based on the adjacency matrix of the subnetwork, remember G 1 The corresponding adjacency matrix is ​​A 1 , remember G 2 The corresponding adjacency matrix is ​​A 2 , and the power iteration spaces corresponding to the two adjacency matrices are Where k is the iteration order, and e is a vector whose components are all 1;

[0062] C. Calculate the similarity of the corresponding sub...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a graph kernel-based social network link prediction method, which belongs to the field of social network analysis. The invention reconstructs the positive link network, the negative link network, or the entire network on the basis of the existing social network, and generates sub-networks under different thresholds for nodes based on the reconstructed network. On this basis, the present invention generates three kinds of sub-network sets of reconstructed networks under different thresholds for nodes, and calculates the similarity between nodes by using the graph kernel method; finally, based on the similarity between nodes, it uses machine learning algorithms to Link Prediction.

Description

technical field [0001] The invention relates to a graph kernel-based social network link prediction method, which belongs to the field of social network analysis. Background technique [0002] With the development of information technology, social network analysis has become a research hotspot in many fields. Social networks are composed of social roles and the connections (+ / -) between roles and roles. Social networks can be regarded as a graph. Social roles It can be seen as a node in the graph, and the connection between roles can be seen as an edge connected to the node. In social network analysis, link prediction is the basis of research [1-4] , because any complex network is derived from the proliferation of simple networks. Link prediction mainly uses the attributes of existing network nodes and the connections between them to predict new links or unknown links that may exist between evaluation nodes. [1，4，5] . As the basic research of social network analysis, lin...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/955G06K9/62G06Q50/00
CPCG06F16/955G06Q50/01G06F18/2411G06F18/24147
Inventor 袁伟伟何康亚李晨亮
Owner NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com